Abstract
Smartphone recycling is a heated topic recently because of environmental and economical concerns. However, the limited recognition ability of naked eyes to find defects on phones impedes further growth of the second-hand markets. Related work for this problem focuses on detection and feature extraction, which can be tough because of the thin shapes of the defects and similar characteristics obstructing discrimination between scratches and cracks.
In this paper, we propose an algorithm using semantic segmentation based on HRNet to build an effective model to evaluate defects of screens quickly and precisely. We improve the performance of the model by introducing the Lovász-Softmax loss function and data augmentation. Peripheral methods including subdivision and postprocessing are proposed to reduce misprediction on small scratches and enforce reliability in practice. We compare our model with other networks on mIoU and Kappa scores, concluding on an empirical basis.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Sarath, P., Bonda, S., Mohanty, S., Nayak, S.K.: Mobile phone waste management and recycling: views and trends. Waste Manage. 46, 536–545 (2015)
Jian, C.X., Gao, J., Chen, X.: A review of TFT-LCD panel defect detection methods. In: Advanced Materials Research, vol. 734, pp. 2898–2902. Trans Tech Publications (2013)
Lei, J., Gao, X., Feng, Z., Qiu, H., Song, M.: Scale insensitive and focus driven mobile screen defect detection in industry. Neurocomputing 294, 72–81 (2018)
Li, C., Zhang, X., Huang, Y., Tang, C., Fatikow, S.: A novel algorithm for defect extraction and classification of mobile phone screen based on machine vision. Comput. Ind. Eng. 146, 106530 (2020)
Ming, W., et al.: A comprehensive review of defect detection in 3C glass components. Measurement 158, 107722 (2020)
Sun, K., Xiao, B., Liu, D., Wang, J.: Deep high-resolution representation learning for human pose estimation. arXiv:1902.09212 (2019)
Sun, K., et al.: High-resolution representations for labeling pixels and regions. arXiv:1904.04514 (2019)
Yu, J., Blaschko, M.: The lovász hinge: a novel convex surrogate for submodular losses. arXiv:1512.07797 (2015)
Lovász, L.: Submodular Functions and Convexity. Springer, Heidelberg, pp. 235–257 (1983). https://doi.org/10.1007/978-3-642-68874-4_10
Berman, M., Triki, A.R., Blaschko, M.B.: The lovász-softmax loss: a tractable surrogate for the optimization of the intersection-over-union measure in neural networks. arXiv:1705.08790 (2017)
Canny, J.: A computational approach to edge detection. IEEE Trans. Pattern Anal. Mach. Intell. PAMI-8(6), 679–698 (1986)
Acknowledgement
This work was supported by the Postdoctoral Science Foundation of China (Grant No. 2019M661510), Science and Technology on Near-Surface Detection Laboratory (Grant No. 6142414180203, 6142414190203), and the Young Scientists Fund of the National Natural Science Foundation of China (Grant No. 41704123).
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zhu, H., Shen, Z., Yuan, S., Chen, F., Li, J. (2021). Semantic Segmentation for Evaluation of Defects on Smartphone Screens. In: Sun, F., Liu, H., Fang, B. (eds) Cognitive Systems and Signal Processing. ICCSIP 2020. Communications in Computer and Information Science, vol 1397. Springer, Singapore. https://doi.org/10.1007/978-981-16-2336-3_41
Download citation
DOI: https://doi.org/10.1007/978-981-16-2336-3_41
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-2335-6
Online ISBN: 978-981-16-2336-3
eBook Packages: Computer ScienceComputer Science (R0)